The application of machine learning in medical diagnosis has gained significant traction due to its potential for early detection and accurate classification of diseases. This study investigates the effectiveness of ten machine learning classifiers—including Decision Tree, Random Forest, Extra Trees, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), in predicting breast cancer and diabetes. Two benchmark datasets were used: the Breast Cancer Wisconsin (Diagnostic) dataset and the Pima Indians Diabetes dataset. Models were evaluated based on Accuracy, Precision, Recall, and F1-Score. The Extra Trees classifier achieved the highest performance on the breast cancer dataset, with an accuracy of 96.49% and an F1-Score of 0.9718. In contrast, performance on the diabetes dataset was more modest, with the Decision Tree achieving the best F1-Score of 0.6549 and an accuracy of 74.68%. These findings highlight the importance of dataset characteristics on model performance and suggest that ensemble methods are particularly effective for structured medical data. Future work should explore advanced preprocessing, feature engineering, and deep learning techniques to enhance prediction in more complex healthcare scenarios.